Patent Application
20210087704
The present disclosure is drawn to electrolytic cells, and in particular, electrolytic cells for surface treatments of discontinuous materials.
Carbon fiber reinforced composites (CFRCs) are used in many commercial applications, including environmental (e.g., wind turbines and blades, fuel cells), aerospace (e.g., air frames), ground vehicles (e.g., luxury car frames, trailers, and car parts), sporting (e.g., tennis rackets, paddles, bicycles, and helmets), medical (e.g., wheel chairs, prosthetic limps, orthotics, in-soles, and braces) and construction (e.g., grid strengthening in precast concrete, foot bridges, and concrete cloth for buildings), and others such as in cell phone frames or watches.
Due to the high strength-to-weight ratio of CFRCs, a growing number of these industries (e.g., environmental, military, aerospace, automobile, sport, construction, and medical industries) are interested in discontinuous carbon fibers for improved formability of CFRCs. Currently, global demand for carbon fibers is ˜60,000 tons/year and the industry is worth at least $3.5 billion. Further demand is expected to grow at an annual rate of 10.9% per year through to 2025.
Because of this growth, it is both environmentally and economically practical to recycle carbon fibers. Recycled carbon fibers can fill the anticipated gap between supply and demand, which would result from the high production costs of carbon fibers. A major barrier to reintroducing recovered and recycled carbon fibers is the lack of an efficient method for surface treating reclaimed chopped fibers that have undergone pyrolysis and other recycling processes that leave the fiber surface essentially without functional groups needed for good fiber-matrix adhesion.
For example, discontinuous fibers cannot be treated using conventional roll-to-roll electrochemical surface treatments that are used for continuous pitch- and poly(acrylonitrile)-based fibers. Because of their relatively fast and uniform treatment, electrochemical methods for discontinuous fibers have already been developed for carbon fiber tows and filaments. However, these methods are not scalable or would damage fibers that are longer than filaments.
A process that can help keep up with carbon fiber demands and encourage the reintroduction of recycled/reclaimed carbon fibers into composites would be commercially and environmentally useful.
A first aspect of the present disclosure is drawn to an electrolytic cell for treating fibers. The cell includes a power supply for providing current. The cell also includes a porous electrically conductive mat within an electrolyte solution, the porous electrically conductive mat being operably connected to the power supply. The mat is configured to allow liquid, gas, or combination thereof to leave an outer surface of a discontinuous material in contact with the porous electrically conductive mat. The cell also includes a porous structure that is configured to keep the discontinuous material in contact with the porous electrically conductive mat while a surface treatment is being applied to the discontinuous fiber.
In some embodiments, the electrolyte solution may comprise water-soluble salts (e.g., ammonia salts, such as ammonium bicarbonate), metal salts (e.g., copper salts, or nickel salts), or a combination thereof. In some embodiments, a voltage of between −1.2V and 1.2V may be applied by the current supply. In some embodiments, the temperature of the electrolyte solution may be between about 3° C. and about 100° C. In some embodiments, the porous electrically conductive mat is configured to function as an electrode. In some embodiments, the polarity of the current is configured to periodically reverse. In some embodiments, the discontinuous material is a carbon fiber. In some embodiments, the porous electrically conductive mat is carbon felt or platinum mesh. In some embodiments, the porous electrically conductive mat is in contact with a porous working electrode. In some embodiments, the porous structure is a platinum or polytetrafluoroethylene (PTFE) mesh. In some embodiments, the porous structure is operably connected to a plunger and is configured to be raised to allow a plurality of discontinuous materials to be added to or removed from the electrolytic cell and lowered to push the plurality of discontinuous materials onto the porous electrically conductive mat.
A second aspect of the present disclosure is drawn to a method for the surface modification of discontinuous conductive materials using, e.g., the electrolytic cell described above. The method begins by introducing a plurality of discontinuous fibers to an electrolyte solution within an electrolytic cell. A porous structure is used to keep the plurality of discontinuous fibers in contact with a porous electrically conductive mat, the porous electrically conductive mat being operably connected to a current supply. The fibers are then treated, by applying current to the electrolytic cell. The fibers can then be removed, or are allowed to be removed, from the porous electrically conductive mat.
The present disclosure is drawn to electrolytic cells for scalable and effective surface treatment of discontinuous carbon fibers. Unlike previous electrochemical surface treatments for small conductive fibers/parts, the present electrochemical cell design (i) does not damage carbon fibers, (ii) is not limited to treating a few carbon-fiber tows in a single run, (iii) provides uniform surface treatment, and (iv) provides modular surface treatment.
These benefits encourage the use of material stamping operations, which are an order of magnitude more cost effective compared to standard composite layup manufacturing.
The disclosed method and apparatus can be understood with reference to
A simplified embodiment of such an electrolytic cell can be seen in reference to
The polarity of the electrolytic cell will often depend on the treatment desired. In some embodiments, the polarity of the current is configured to reverse at some point in time during a surface treatment process, typically as either a step change, or alternating in a periodic fashion. For example, in some embodiments, the system spends 5 minutes at a first polarity, then spends 5 minutes at the reverse polarity.
As can be seen in
The solvent in the electrolyte solution may be any appropriate solvent. In some embodiments, the solvent is water. In some embodiments, the solvent is non-aqueous, i.e., substantially free of water.
Although not shown in
As can be seen in
The porous electrically conductive mat (220) is operably connected to the power supply (210).
The porous electrically conductive mat may be comprised of one or more materials. For example, the mat may be a non-woven fiber substrate, where the fibers are coated with, e.g., and electrically conductive material. In some embodiments, the mat comprises treated or untreated carbon felt, or a platinum mesh.
Typically, the average pore size of the mat is sufficiently small that the discontinuous material cannot readily pass through the mat. That is, the average pore size is typically smaller than the smallest dimension of the discontinuous material to be treated (e.g., for fibers, the smallest dimension is the diameter, not the length, while for a spherical particle, the smallest dimension is the diameter.).
The porous electrically conductive mat is configured to allow liquid, gas, or combination thereof to leave an outer surface of a discontinuous material (230) that is in contact with the porous electrically conductive mat (220) during normal operation of the electrolytic cell. That is, gases and liquids must be substantially free to exit the surface of the discontinuous material (230) that are within the electrolyte and being treated.
The discontinuous material (230) can be any electrically conductive discontinuous material that requires surface treatment. The material may be in any form (particle, fiber, etc.). Preferred embodiments of discontinuous materials are conductive non-metal fibers, such as graphene or carbon fibers, or metal or metal alloy fibers, including, e.g., aluminum or copper fibers.
The discontinuous material may be kept in contact with the porous electrically conductive mat via an applied mechanical force, gravity, or fluid flow. As seen in
The porous structure can be comprised of any appropriate material. In some embodiments, the porous structure is a platinum or polytetrafluoroethylene (PTFE) mesh. In some embodiments, the porous structure is a substantially rigid structure. In some embodiments, the porous structure is a flexible woven or non-woven material, such as a paper towel, cheesecloth, or wool.
Referring briefly to
Referring briefly to
Referring briefly to
Referring briefly to
The disclosed cells may contain other components, including but not limited to, additional valves, sensors, additional circuitry, and one or more processors configured to control the process. These additional components are typically present to help automate, control, or monitor the process. For example, in one embodiment, the cell utilizes a camera to visually monitor the porous conductive mat and ensure it is visually in good condition.
Referring back to
The discontinuous materials can be introduced into the electrolyte within the electrolytic cell in any known manner, including, but not limited to, batch addition of the discontinuous conductive materials directly into a cell containing the electrolyte, or premixing the fibers and the electrolyte, and using a valve to allow the premix to flow into the cell periodically or continuously.
Once the materials are introduced, the porous structure can be used (130) to keep the plurality of discontinuous materials in contact with the porous electrically conductive mat. As disclosed above, this may be done by any appropriate means, including, e.g., directly, by pressing the discontinuous materials onto the mat, or indirectly, by preventing the flow of the electrolyte solution from leaving the surface of the mat.
With the discontinuous materials in contact with the porous electrically conducting mat, the discontinuous materials can be treating by applying current to the electrolytic cell for a period of time. In some embodiments, the treatment time is less than 20 minutes. In some embodiments, the treatment time is less than 10 minutes. In some embodiments, the treatment time is greater than 1 minute. In some embodiments, the treatment time is greater than 5 minutes.
The present disclosure provides broad control over surface chemistry, and settings can be adjusted such that the results are similar to those achieved via conventional continuous fiber surface treatments.
Following treatment, the fibers are removed (e.g., the porous conductive mat can be removed, scraped, etc.), or allowed to be removed (e.g., lifting the non-conductive porous structure off the mat, flushing new electrolyte through the cell, ensuring the fibers are lifted off the mat and out through an outlet port).
Untreated, de-sized discontinuous carbon fibers (pitch-based, K223HE) with lengths 6 mm and 25 mm were obtained from Mitsubishi Chemical Carbon Fiber and Composites. A platinum mesh (2.54 cm×2.54 cm) electrode was purchased from Ametek Scientific Instruments, Mueller BU-60C alligator clips were obtained from McMaster-Carr, and a porous electrically conductive mat (carbon felt) was purchased from Metaullics Systems Company, L.P. Electrolyte solutions were prepared from Milli-Q water (18 MΩ·cm−1) and ammonium bicarbonate (>99.5%, Sigma Aldrich), ammonium chloride (>99.5%, Sigma Aldrich), ammonium hydroxide solution (28 wt % ammonia in water, Sigma Aldrich), ammonium carbamate (99%, Sigma Aldrich) or sodium bicarbonate (>99.7%, Sigma Aldrich). In order to obtain a snug fitting mesh for specific chemical glasswares (i.e., the porous structure) and have greater flexibility over open areas (or pore size) in the mesh, the PTFE meshes used in this study were made by rolling up and weaving polytetrafluroethylene thread seal tape (ULINE, 12.7 mm width×0.07 pm thickness). However, commercially available PTFE meshes also worked.
Electrochemical surface treatments were performed in an undivided cell, similar to the one shown in
X-ray photoelectron spectroscopy (XPS) data were collected on a Physical Electronics Versa Probe II instrument with an Al Kα source. Compositions were obtained from high-resolution scans with a pass energy of 23.5 eV and a step size of 0.05 eV for the C 1s, O 1s, and N 1s regions. CasaXPS software was used to process XPS data. For each experiment, a minimum of three randomly selected carbon fiber bundles were analyzed. Scanning electron microscopy (SEM) was performed on a Hitachi S-4700 at an accelerating voltage of 5 kV. Prior to imaging, samples were sputtered with an approximately 2 nm thick coating of iridium.
Single fiber tensile tests were performed on 25 mm long carbon fibers using an automatic single fiber test system called a Favimat (Textechno H. Stein GmbH & Co. KG, Germany). The load cell capacity was 210 cN, the force resolution was 0.0001 cN, the maximum possible travel was 100 mm, and the elongation resolution was 0.1 pm. The Favimat performed an independent measurement of the fiber linear density using an acoustic resonance technique prior to each individual tensile test. The diameter of the fiber being tested was obtained from this measurement by assuming that the fiber has a cylindrical shape. The experiments were performed using grips supplied with the Favimat that are designed to grip carbon fibers between surfaces of hard and soft rubber. Testing parameters include a pretension of 0.5 cN/tex, a displacement rate of 0.9 mm/min, and a gauge length of 15 mm. Single fibers were loaded directly into the Favimat grips without the assistance of paper or cardstock tabbing. Eleven tests are reported for the untreated fibers, and twenty-one tests are reported for treated fibers. Thirty tests (with no slipping) were reported each for untreated and treated fibers. Physical properties, including diameter, break strength, and initial modulus, were collected from the software program included with the Favimat, Textechno35 (Version 3.0.3547).
Conductivity and pH measurements were performed on the Oakton™ PC 700 instruments. The pH probe was calibrated no more than a week prior to measurement, the conductivity probe was calibrated according to the manufacturer's specifications using standard calibration solutions purchased from Oakton. The pH values and conductivities of the most commonly used electrolyte solution (i.e., aqueous ammonium bicarbonate or NH4HCO3), which was used in this work, are given in Table 1 below.
Proof-of-concept for the electrochemical treatment of discontinuous fibers was confirmed by XPS analysis. Because the surface of as-received dialead fibers consisted of 98.2-99.6 at. % (atomic %) carbon and 0.4-1.3 at. % oxygen, either an increase in surface oxygen or the presence of surface nitrogen would indicate successful electrochemical oxidation with ammonium bicarbonate. Using preliminary electrochemical parameters (i.e., applied voltage=16 V, 10 min. anodic oxidation, starting temperature ˜23° C., and electrolyte concentration=0.50 M), the surface oxygen content increased to 6.5+0.4 at. %, and the surface nitrogen content was 6.7+0.3 at. % (See
Electrochemical processing conditions can be adjusted or optimized to achieve, e.g., surface functionalization similar to commercially available electrochemically treated continuous fibers.
To determine which applied voltage leads to the highest loading of surface functional groups without damaging the carbon fibers, 10 min. anodic oxidation was performed with varying applied voltages (1.5V to 21V) in 0.5 M ammonium bicarbonate (aq.) at an initial temperature of ˜23° C. In this example, the at. % of surface nitrogen increases steadily with increasing applied voltage, while that of oxygen remains relatively constant above a bias of 1.5 V. Hence, the oxygen-to-nitrogen (O:N) ratio changes over the 1.5-21 V range from more oxygen than nitrogen to equal amounts and then to more nitrogen than oxygen. At higher applied voltages (>14 V), the atomic % of surface heteroatoms (or non-carbon and non-hydrogen atoms) does not change.
Based on the SEM micrographs, no obvious surface damage was observed until 21 V at 35 k magnification. That is, no pitting or deepening of striations were observed for fibers treated with lower applied voltages when other parameters (i.e., starting temperature ˜23° C., electrolyte concentration=0.50 M, and 10 min. at anodic oxidation only) were held constant. However, both voltage and time play a role in determining damage done to the discontinuous material; interestingly, long anodic oxidation (2 h) with relatively low voltage (i.e., 1.5 V) causes severe pitting and striations that penetrate deep into the internal structure of the carbon fibers. Because good surface functionalization without obvious fiber damage were observed by XPS and SEM at 16 V, this applied voltage was used in further optimizations discussed below.
Carbon fibers are capable of both anodic oxidation and cathodic reduction. Referring to Table 2, it is seen that broad control of the nitrogen and oxygen contents can be achieved by changing electrode polarity and polarity time. The fibers are subjected to an applied voltage of 16 V, an initial temperature of ˜23° C., and ammonium bicarbonate concentration of 0.5 M. In all cases, the anodic treatment (if any) was conducted before the cathodic treatment. Treatments at both the anode and the cathode increase surface oxygen and nitrogen. Slightly more nitrogen is observed under only anodic oxidation (Table 2, row 1), and the opposite is true for cathodic reduction only (Table 2, row 5). Total amount of heteroatoms peaked at 80% anode (or 8 min at the anode followed by 2 min at cathode). Similar to the case for applied voltage (above), the O:N ratio decreases with increasing time at the anode. However, the surface never contained significantly more nitrogen than oxygen. When ammonium bicarbonate is the only electrolyte, the electrochemical procedure that best resembles the commercially available fibers is 5 min at anodic oxidation, followed by 5 min at cathodic reduction for discontinuous carbon fibers (Table 3). This mix-biased procedure was used for the remainder of the ammonium bicarbonate studies.
To determine the reproducibility of the electrochemical cell design, four separate experiments were performed, and XPS was conducted on 3-12 randomly chosen fibers in each experiment. Low standard deviations in the overall atomic concentrations were observed in each of the four separate experiments (Table 4, rows 1-4) as well as in the combined 21 replications (Table 4, row 5). Furthermore, peak fitting was performed for the C is region and noted that there was good consistency in these elemental speciation results (Table 5). Therefore, the disclosed electrolytic method is highly reproducible and relatively uniform for the surface treatment of short carbon fibers.
Broad control over surface loading of nitrogen and oxygen can also be achieved by changing the initial temperature of electrochemical reaction. Experiments were completed non-isothermally, with starting temperatures running from 3° C. to about 70.5° C. Temperatures increased by as much as 40° C. over the course of a run. When fibers were subjected to 5 min at anodic oxidation followed by 5 min at cathodic reduction with an applied voltage of 16 V in an ammonium bicarbonate solution (aq. 0.5 M), nitrogen content varied linearly with initial temperature, from about 3 at. % to about 9 at. %. On the other hand, the oxygen content did not change much, varying between about 7 at. % and 9 at. %, and only a small decrease (to about 6 at. %) was observed at initial temperature of 70.5° C. Similar to the case for applied voltage (above), the carbon fiber surface changed from oxygen-rich to nitrogen-rich. This change occurs around the initial temperature of 50° C., which is also the initial temperature that leads to the highest total amount of heteroatoms.
Another processing variable that significantly affects the surface loading of nitrogen and oxygen is electrolyte concentration. When discontinuous fibers were subjected to 5 min anodic oxidation followed by 5 min cathodic reduction with an applied voltage of 16 V and an initial temperature of ˜23° C., both nitrogen and oxygen loadings increased steadily with electrolyte concentration (here, tested from 0.005M to 0.75M). At 0.005M, significant surface O content (about 4 at. %) but negligible N content were observed. Even 0.1M concentrations see a large jump in loadings (O content about 7 at. %, N content about 2 at. %). At 0.75 M, the oxygen (about 10.5 at. %) and nitrogen (about 5 at. %) contents are within the standard deviations of commercially available, electrochemically treated continuous carbon fibers. Unlike the cases for applied voltage and initial temperature, the surface remains oxygen-rich at all concentrations.
To confirm the electrochemical method does not adversely affect tensile properties of carbon fibers, single fiber tensile tests were performed. The tensile strength for untreated fibers and treated fibers were very similar, 3395±996 and 3691±1500 MPa, respectively. Here, treated fibers were treated for 5 min at the anode fallowed by 5 min at the cathode with an applied voltage of 16 V, an ammonium bicarbonate concentration of 0.5 M, and an initial temperature of ˜22° C.) The tensile moduli of the untreated and treated carbon fibers are also very similar (833±214 and 826±104). Due to the inherently brittle nature of carbon fiber, it is typical of single fiber testing to observe relatively high standard deviations. Nevertheless, the single fiber tensile tests indicate that the disclosed electrochemical method does not significantly change the tensile properties of the untreated fibers. Based on the manufacturer's spec sheet, the as-received chopped carbon fiber tows have a tensile strength of 3800 MPa and a tensile modulus of 900 GPa. While these mechanical data are slightly higher than the results in this experiment, this difference is probably due to the manufacturers data being based on fiber tow tests, while our tensile tests are based on single fibers. However, the manufacturer's data are still within the statistical error of the tensile results.
Those skilled in the art will recognize or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.
This application claims priority benefit to U.S. Provisional Patent Application No. 62/905,578, filed on Sep. 25, 2019, which is incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 62905578 | Sep 2019 | US |
